File size: 24,359 Bytes
565bafc |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 |
2023-10-19 20:58:16,394 ----------------------------------------------------------------------------------------------------
2023-10-19 20:58:16,394 Model: "SequenceTagger(
(embeddings): TransformerWordEmbeddings(
(model): BertModel(
(embeddings): BertEmbeddings(
(word_embeddings): Embedding(32001, 128)
(position_embeddings): Embedding(512, 128)
(token_type_embeddings): Embedding(2, 128)
(LayerNorm): LayerNorm((128,), eps=1e-12, elementwise_affine=True)
(dropout): Dropout(p=0.1, inplace=False)
)
(encoder): BertEncoder(
(layer): ModuleList(
(0-1): 2 x BertLayer(
(attention): BertAttention(
(self): BertSelfAttention(
(query): Linear(in_features=128, out_features=128, bias=True)
(key): Linear(in_features=128, out_features=128, bias=True)
(value): Linear(in_features=128, out_features=128, bias=True)
(dropout): Dropout(p=0.1, inplace=False)
)
(output): BertSelfOutput(
(dense): Linear(in_features=128, out_features=128, bias=True)
(LayerNorm): LayerNorm((128,), eps=1e-12, elementwise_affine=True)
(dropout): Dropout(p=0.1, inplace=False)
)
)
(intermediate): BertIntermediate(
(dense): Linear(in_features=128, out_features=512, bias=True)
(intermediate_act_fn): GELUActivation()
)
(output): BertOutput(
(dense): Linear(in_features=512, out_features=128, bias=True)
(LayerNorm): LayerNorm((128,), eps=1e-12, elementwise_affine=True)
(dropout): Dropout(p=0.1, inplace=False)
)
)
)
)
(pooler): BertPooler(
(dense): Linear(in_features=128, out_features=128, bias=True)
(activation): Tanh()
)
)
)
(locked_dropout): LockedDropout(p=0.5)
(linear): Linear(in_features=128, out_features=17, bias=True)
(loss_function): CrossEntropyLoss()
)"
2023-10-19 20:58:16,394 ----------------------------------------------------------------------------------------------------
2023-10-19 20:58:16,394 MultiCorpus: 7142 train + 698 dev + 2570 test sentences
- NER_HIPE_2022 Corpus: 7142 train + 698 dev + 2570 test sentences - /root/.flair/datasets/ner_hipe_2022/v2.1/newseye/fr/with_doc_seperator
2023-10-19 20:58:16,394 ----------------------------------------------------------------------------------------------------
2023-10-19 20:58:16,394 Train: 7142 sentences
2023-10-19 20:58:16,394 (train_with_dev=False, train_with_test=False)
2023-10-19 20:58:16,394 ----------------------------------------------------------------------------------------------------
2023-10-19 20:58:16,395 Training Params:
2023-10-19 20:58:16,395 - learning_rate: "5e-05"
2023-10-19 20:58:16,395 - mini_batch_size: "4"
2023-10-19 20:58:16,395 - max_epochs: "10"
2023-10-19 20:58:16,395 - shuffle: "True"
2023-10-19 20:58:16,395 ----------------------------------------------------------------------------------------------------
2023-10-19 20:58:16,395 Plugins:
2023-10-19 20:58:16,395 - TensorboardLogger
2023-10-19 20:58:16,395 - LinearScheduler | warmup_fraction: '0.1'
2023-10-19 20:58:16,395 ----------------------------------------------------------------------------------------------------
2023-10-19 20:58:16,395 Final evaluation on model from best epoch (best-model.pt)
2023-10-19 20:58:16,395 - metric: "('micro avg', 'f1-score')"
2023-10-19 20:58:16,395 ----------------------------------------------------------------------------------------------------
2023-10-19 20:58:16,395 Computation:
2023-10-19 20:58:16,395 - compute on device: cuda:0
2023-10-19 20:58:16,395 - embedding storage: none
2023-10-19 20:58:16,395 ----------------------------------------------------------------------------------------------------
2023-10-19 20:58:16,395 Model training base path: "hmbench-newseye/fr-dbmdz/bert-tiny-historic-multilingual-cased-bs4-wsFalse-e10-lr5e-05-poolingfirst-layers-1-crfFalse-5"
2023-10-19 20:58:16,395 ----------------------------------------------------------------------------------------------------
2023-10-19 20:58:16,395 ----------------------------------------------------------------------------------------------------
2023-10-19 20:58:16,395 Logging anything other than scalars to TensorBoard is currently not supported.
2023-10-19 20:58:19,025 epoch 1 - iter 178/1786 - loss 3.39433925 - time (sec): 2.63 - samples/sec: 9379.21 - lr: 0.000005 - momentum: 0.000000
2023-10-19 20:58:21,707 epoch 1 - iter 356/1786 - loss 2.91442988 - time (sec): 5.31 - samples/sec: 9619.01 - lr: 0.000010 - momentum: 0.000000
2023-10-19 20:58:24,768 epoch 1 - iter 534/1786 - loss 2.29912039 - time (sec): 8.37 - samples/sec: 9185.83 - lr: 0.000015 - momentum: 0.000000
2023-10-19 20:58:27,960 epoch 1 - iter 712/1786 - loss 1.91163547 - time (sec): 11.56 - samples/sec: 8856.81 - lr: 0.000020 - momentum: 0.000000
2023-10-19 20:58:31,072 epoch 1 - iter 890/1786 - loss 1.69293519 - time (sec): 14.68 - samples/sec: 8590.79 - lr: 0.000025 - momentum: 0.000000
2023-10-19 20:58:34,159 epoch 1 - iter 1068/1786 - loss 1.52968510 - time (sec): 17.76 - samples/sec: 8452.84 - lr: 0.000030 - momentum: 0.000000
2023-10-19 20:58:37,316 epoch 1 - iter 1246/1786 - loss 1.39348187 - time (sec): 20.92 - samples/sec: 8397.32 - lr: 0.000035 - momentum: 0.000000
2023-10-19 20:58:40,371 epoch 1 - iter 1424/1786 - loss 1.28507688 - time (sec): 23.98 - samples/sec: 8375.29 - lr: 0.000040 - momentum: 0.000000
2023-10-19 20:58:43,370 epoch 1 - iter 1602/1786 - loss 1.19861795 - time (sec): 26.97 - samples/sec: 8325.77 - lr: 0.000045 - momentum: 0.000000
2023-10-19 20:58:46,435 epoch 1 - iter 1780/1786 - loss 1.13141747 - time (sec): 30.04 - samples/sec: 8257.10 - lr: 0.000050 - momentum: 0.000000
2023-10-19 20:58:46,527 ----------------------------------------------------------------------------------------------------
2023-10-19 20:58:46,527 EPOCH 1 done: loss 1.1299 - lr: 0.000050
2023-10-19 20:58:47,918 DEV : loss 0.3072870969772339 - f1-score (micro avg) 0.2262
2023-10-19 20:58:47,932 saving best model
2023-10-19 20:58:47,966 ----------------------------------------------------------------------------------------------------
2023-10-19 20:58:51,058 epoch 2 - iter 178/1786 - loss 0.46644868 - time (sec): 3.09 - samples/sec: 7639.57 - lr: 0.000049 - momentum: 0.000000
2023-10-19 20:58:54,074 epoch 2 - iter 356/1786 - loss 0.46322785 - time (sec): 6.11 - samples/sec: 7992.93 - lr: 0.000049 - momentum: 0.000000
2023-10-19 20:58:57,117 epoch 2 - iter 534/1786 - loss 0.45211784 - time (sec): 9.15 - samples/sec: 8006.32 - lr: 0.000048 - momentum: 0.000000
2023-10-19 20:59:00,061 epoch 2 - iter 712/1786 - loss 0.44022463 - time (sec): 12.09 - samples/sec: 8050.83 - lr: 0.000048 - momentum: 0.000000
2023-10-19 20:59:03,120 epoch 2 - iter 890/1786 - loss 0.42492323 - time (sec): 15.15 - samples/sec: 8078.68 - lr: 0.000047 - momentum: 0.000000
2023-10-19 20:59:06,209 epoch 2 - iter 1068/1786 - loss 0.41906923 - time (sec): 18.24 - samples/sec: 8122.04 - lr: 0.000047 - momentum: 0.000000
2023-10-19 20:59:09,470 epoch 2 - iter 1246/1786 - loss 0.41285425 - time (sec): 21.50 - samples/sec: 8094.77 - lr: 0.000046 - momentum: 0.000000
2023-10-19 20:59:12,467 epoch 2 - iter 1424/1786 - loss 0.41328103 - time (sec): 24.50 - samples/sec: 8047.76 - lr: 0.000046 - momentum: 0.000000
2023-10-19 20:59:15,495 epoch 2 - iter 1602/1786 - loss 0.40602105 - time (sec): 27.53 - samples/sec: 8059.81 - lr: 0.000045 - momentum: 0.000000
2023-10-19 20:59:18,579 epoch 2 - iter 1780/1786 - loss 0.40352887 - time (sec): 30.61 - samples/sec: 8101.49 - lr: 0.000044 - momentum: 0.000000
2023-10-19 20:59:18,683 ----------------------------------------------------------------------------------------------------
2023-10-19 20:59:18,683 EPOCH 2 done: loss 0.4037 - lr: 0.000044
2023-10-19 20:59:21,544 DEV : loss 0.23947705328464508 - f1-score (micro avg) 0.3628
2023-10-19 20:59:21,558 saving best model
2023-10-19 20:59:21,594 ----------------------------------------------------------------------------------------------------
2023-10-19 20:59:24,722 epoch 3 - iter 178/1786 - loss 0.30638540 - time (sec): 3.13 - samples/sec: 8248.30 - lr: 0.000044 - momentum: 0.000000
2023-10-19 20:59:27,738 epoch 3 - iter 356/1786 - loss 0.32648450 - time (sec): 6.14 - samples/sec: 8067.28 - lr: 0.000043 - momentum: 0.000000
2023-10-19 20:59:30,745 epoch 3 - iter 534/1786 - loss 0.33121133 - time (sec): 9.15 - samples/sec: 8004.37 - lr: 0.000043 - momentum: 0.000000
2023-10-19 20:59:33,798 epoch 3 - iter 712/1786 - loss 0.32647076 - time (sec): 12.20 - samples/sec: 8063.71 - lr: 0.000042 - momentum: 0.000000
2023-10-19 20:59:36,949 epoch 3 - iter 890/1786 - loss 0.32878218 - time (sec): 15.35 - samples/sec: 8065.26 - lr: 0.000042 - momentum: 0.000000
2023-10-19 20:59:40,305 epoch 3 - iter 1068/1786 - loss 0.33069697 - time (sec): 18.71 - samples/sec: 7989.44 - lr: 0.000041 - momentum: 0.000000
2023-10-19 20:59:43,433 epoch 3 - iter 1246/1786 - loss 0.32906811 - time (sec): 21.84 - samples/sec: 7995.43 - lr: 0.000041 - momentum: 0.000000
2023-10-19 20:59:46,484 epoch 3 - iter 1424/1786 - loss 0.32966262 - time (sec): 24.89 - samples/sec: 7989.64 - lr: 0.000040 - momentum: 0.000000
2023-10-19 20:59:49,454 epoch 3 - iter 1602/1786 - loss 0.32998018 - time (sec): 27.86 - samples/sec: 7989.40 - lr: 0.000039 - momentum: 0.000000
2023-10-19 20:59:52,398 epoch 3 - iter 1780/1786 - loss 0.32830930 - time (sec): 30.80 - samples/sec: 8052.42 - lr: 0.000039 - momentum: 0.000000
2023-10-19 20:59:52,479 ----------------------------------------------------------------------------------------------------
2023-10-19 20:59:52,479 EPOCH 3 done: loss 0.3291 - lr: 0.000039
2023-10-19 20:59:54,832 DEV : loss 0.21711337566375732 - f1-score (micro avg) 0.4412
2023-10-19 20:59:54,845 saving best model
2023-10-19 20:59:54,880 ----------------------------------------------------------------------------------------------------
2023-10-19 20:59:58,253 epoch 4 - iter 178/1786 - loss 0.28910294 - time (sec): 3.37 - samples/sec: 7599.18 - lr: 0.000038 - momentum: 0.000000
2023-10-19 21:00:01,330 epoch 4 - iter 356/1786 - loss 0.28554259 - time (sec): 6.45 - samples/sec: 7654.04 - lr: 0.000038 - momentum: 0.000000
2023-10-19 21:00:04,474 epoch 4 - iter 534/1786 - loss 0.28752177 - time (sec): 9.59 - samples/sec: 7716.32 - lr: 0.000037 - momentum: 0.000000
2023-10-19 21:00:07,494 epoch 4 - iter 712/1786 - loss 0.29481804 - time (sec): 12.61 - samples/sec: 7746.74 - lr: 0.000037 - momentum: 0.000000
2023-10-19 21:00:10,542 epoch 4 - iter 890/1786 - loss 0.29209292 - time (sec): 15.66 - samples/sec: 7825.10 - lr: 0.000036 - momentum: 0.000000
2023-10-19 21:00:13,639 epoch 4 - iter 1068/1786 - loss 0.29321400 - time (sec): 18.76 - samples/sec: 7799.47 - lr: 0.000036 - momentum: 0.000000
2023-10-19 21:00:16,855 epoch 4 - iter 1246/1786 - loss 0.29088719 - time (sec): 21.97 - samples/sec: 7788.31 - lr: 0.000035 - momentum: 0.000000
2023-10-19 21:00:20,030 epoch 4 - iter 1424/1786 - loss 0.28877714 - time (sec): 25.15 - samples/sec: 7893.11 - lr: 0.000034 - momentum: 0.000000
2023-10-19 21:00:23,060 epoch 4 - iter 1602/1786 - loss 0.28795049 - time (sec): 28.18 - samples/sec: 7909.66 - lr: 0.000034 - momentum: 0.000000
2023-10-19 21:00:26,192 epoch 4 - iter 1780/1786 - loss 0.28608450 - time (sec): 31.31 - samples/sec: 7908.48 - lr: 0.000033 - momentum: 0.000000
2023-10-19 21:00:26,298 ----------------------------------------------------------------------------------------------------
2023-10-19 21:00:26,298 EPOCH 4 done: loss 0.2856 - lr: 0.000033
2023-10-19 21:00:29,109 DEV : loss 0.21065032482147217 - f1-score (micro avg) 0.4851
2023-10-19 21:00:29,122 saving best model
2023-10-19 21:00:29,158 ----------------------------------------------------------------------------------------------------
2023-10-19 21:00:32,181 epoch 5 - iter 178/1786 - loss 0.26936022 - time (sec): 3.02 - samples/sec: 7810.87 - lr: 0.000033 - momentum: 0.000000
2023-10-19 21:00:35,233 epoch 5 - iter 356/1786 - loss 0.25893596 - time (sec): 6.07 - samples/sec: 8113.38 - lr: 0.000032 - momentum: 0.000000
2023-10-19 21:00:38,268 epoch 5 - iter 534/1786 - loss 0.26522534 - time (sec): 9.11 - samples/sec: 8061.37 - lr: 0.000032 - momentum: 0.000000
2023-10-19 21:00:41,439 epoch 5 - iter 712/1786 - loss 0.26176498 - time (sec): 12.28 - samples/sec: 8024.11 - lr: 0.000031 - momentum: 0.000000
2023-10-19 21:00:44,487 epoch 5 - iter 890/1786 - loss 0.26297763 - time (sec): 15.33 - samples/sec: 8044.77 - lr: 0.000031 - momentum: 0.000000
2023-10-19 21:00:47,482 epoch 5 - iter 1068/1786 - loss 0.25919324 - time (sec): 18.32 - samples/sec: 8016.32 - lr: 0.000030 - momentum: 0.000000
2023-10-19 21:00:50,611 epoch 5 - iter 1246/1786 - loss 0.25806711 - time (sec): 21.45 - samples/sec: 8119.10 - lr: 0.000029 - momentum: 0.000000
2023-10-19 21:00:53,748 epoch 5 - iter 1424/1786 - loss 0.25466980 - time (sec): 24.59 - samples/sec: 8072.99 - lr: 0.000029 - momentum: 0.000000
2023-10-19 21:00:56,856 epoch 5 - iter 1602/1786 - loss 0.25598645 - time (sec): 27.70 - samples/sec: 8070.63 - lr: 0.000028 - momentum: 0.000000
2023-10-19 21:00:59,939 epoch 5 - iter 1780/1786 - loss 0.25465192 - time (sec): 30.78 - samples/sec: 8048.62 - lr: 0.000028 - momentum: 0.000000
2023-10-19 21:01:00,052 ----------------------------------------------------------------------------------------------------
2023-10-19 21:01:00,052 EPOCH 5 done: loss 0.2543 - lr: 0.000028
2023-10-19 21:01:02,433 DEV : loss 0.20281952619552612 - f1-score (micro avg) 0.4948
2023-10-19 21:01:02,448 saving best model
2023-10-19 21:01:02,483 ----------------------------------------------------------------------------------------------------
2023-10-19 21:01:05,510 epoch 6 - iter 178/1786 - loss 0.23393227 - time (sec): 3.03 - samples/sec: 7808.95 - lr: 0.000027 - momentum: 0.000000
2023-10-19 21:01:08,545 epoch 6 - iter 356/1786 - loss 0.23081599 - time (sec): 6.06 - samples/sec: 8163.67 - lr: 0.000027 - momentum: 0.000000
2023-10-19 21:01:11,648 epoch 6 - iter 534/1786 - loss 0.23479214 - time (sec): 9.16 - samples/sec: 8155.13 - lr: 0.000026 - momentum: 0.000000
2023-10-19 21:01:14,701 epoch 6 - iter 712/1786 - loss 0.23298850 - time (sec): 12.22 - samples/sec: 8207.54 - lr: 0.000026 - momentum: 0.000000
2023-10-19 21:01:17,670 epoch 6 - iter 890/1786 - loss 0.23743133 - time (sec): 15.19 - samples/sec: 8100.39 - lr: 0.000025 - momentum: 0.000000
2023-10-19 21:01:20,467 epoch 6 - iter 1068/1786 - loss 0.23775517 - time (sec): 17.98 - samples/sec: 8153.61 - lr: 0.000024 - momentum: 0.000000
2023-10-19 21:01:23,606 epoch 6 - iter 1246/1786 - loss 0.23641985 - time (sec): 21.12 - samples/sec: 8149.38 - lr: 0.000024 - momentum: 0.000000
2023-10-19 21:01:26,829 epoch 6 - iter 1424/1786 - loss 0.23416317 - time (sec): 24.35 - samples/sec: 8123.71 - lr: 0.000023 - momentum: 0.000000
2023-10-19 21:01:29,978 epoch 6 - iter 1602/1786 - loss 0.23280312 - time (sec): 27.49 - samples/sec: 8121.44 - lr: 0.000023 - momentum: 0.000000
2023-10-19 21:01:33,180 epoch 6 - iter 1780/1786 - loss 0.23351653 - time (sec): 30.70 - samples/sec: 8085.38 - lr: 0.000022 - momentum: 0.000000
2023-10-19 21:01:33,290 ----------------------------------------------------------------------------------------------------
2023-10-19 21:01:33,291 EPOCH 6 done: loss 0.2338 - lr: 0.000022
2023-10-19 21:01:36,137 DEV : loss 0.2004702240228653 - f1-score (micro avg) 0.5138
2023-10-19 21:01:36,151 saving best model
2023-10-19 21:01:36,187 ----------------------------------------------------------------------------------------------------
2023-10-19 21:01:39,079 epoch 7 - iter 178/1786 - loss 0.23612974 - time (sec): 2.89 - samples/sec: 8198.00 - lr: 0.000022 - momentum: 0.000000
2023-10-19 21:01:42,193 epoch 7 - iter 356/1786 - loss 0.21613429 - time (sec): 6.01 - samples/sec: 8097.03 - lr: 0.000021 - momentum: 0.000000
2023-10-19 21:01:45,222 epoch 7 - iter 534/1786 - loss 0.21090820 - time (sec): 9.03 - samples/sec: 8043.86 - lr: 0.000021 - momentum: 0.000000
2023-10-19 21:01:48,279 epoch 7 - iter 712/1786 - loss 0.21151190 - time (sec): 12.09 - samples/sec: 8048.66 - lr: 0.000020 - momentum: 0.000000
2023-10-19 21:01:51,326 epoch 7 - iter 890/1786 - loss 0.21415014 - time (sec): 15.14 - samples/sec: 8097.58 - lr: 0.000019 - momentum: 0.000000
2023-10-19 21:01:54,448 epoch 7 - iter 1068/1786 - loss 0.22230949 - time (sec): 18.26 - samples/sec: 8094.44 - lr: 0.000019 - momentum: 0.000000
2023-10-19 21:01:57,518 epoch 7 - iter 1246/1786 - loss 0.22203000 - time (sec): 21.33 - samples/sec: 8038.74 - lr: 0.000018 - momentum: 0.000000
2023-10-19 21:02:00,602 epoch 7 - iter 1424/1786 - loss 0.22048490 - time (sec): 24.41 - samples/sec: 8093.24 - lr: 0.000018 - momentum: 0.000000
2023-10-19 21:02:03,686 epoch 7 - iter 1602/1786 - loss 0.21897496 - time (sec): 27.50 - samples/sec: 8107.85 - lr: 0.000017 - momentum: 0.000000
2023-10-19 21:02:06,657 epoch 7 - iter 1780/1786 - loss 0.21764711 - time (sec): 30.47 - samples/sec: 8128.14 - lr: 0.000017 - momentum: 0.000000
2023-10-19 21:02:06,760 ----------------------------------------------------------------------------------------------------
2023-10-19 21:02:06,761 EPOCH 7 done: loss 0.2174 - lr: 0.000017
2023-10-19 21:02:09,120 DEV : loss 0.20184864103794098 - f1-score (micro avg) 0.5363
2023-10-19 21:02:09,133 saving best model
2023-10-19 21:02:09,167 ----------------------------------------------------------------------------------------------------
2023-10-19 21:02:12,237 epoch 8 - iter 178/1786 - loss 0.19631205 - time (sec): 3.07 - samples/sec: 8420.53 - lr: 0.000016 - momentum: 0.000000
2023-10-19 21:02:15,400 epoch 8 - iter 356/1786 - loss 0.20351488 - time (sec): 6.23 - samples/sec: 8448.43 - lr: 0.000016 - momentum: 0.000000
2023-10-19 21:02:18,524 epoch 8 - iter 534/1786 - loss 0.19623338 - time (sec): 9.36 - samples/sec: 8444.47 - lr: 0.000015 - momentum: 0.000000
2023-10-19 21:02:21,610 epoch 8 - iter 712/1786 - loss 0.20086781 - time (sec): 12.44 - samples/sec: 8326.28 - lr: 0.000014 - momentum: 0.000000
2023-10-19 21:02:24,651 epoch 8 - iter 890/1786 - loss 0.20639441 - time (sec): 15.48 - samples/sec: 8245.95 - lr: 0.000014 - momentum: 0.000000
2023-10-19 21:02:27,907 epoch 8 - iter 1068/1786 - loss 0.20336709 - time (sec): 18.74 - samples/sec: 8209.92 - lr: 0.000013 - momentum: 0.000000
2023-10-19 21:02:30,983 epoch 8 - iter 1246/1786 - loss 0.20354696 - time (sec): 21.81 - samples/sec: 8112.38 - lr: 0.000013 - momentum: 0.000000
2023-10-19 21:02:34,043 epoch 8 - iter 1424/1786 - loss 0.20314656 - time (sec): 24.88 - samples/sec: 8074.10 - lr: 0.000012 - momentum: 0.000000
2023-10-19 21:02:37,160 epoch 8 - iter 1602/1786 - loss 0.20316820 - time (sec): 27.99 - samples/sec: 8039.11 - lr: 0.000012 - momentum: 0.000000
2023-10-19 21:02:40,110 epoch 8 - iter 1780/1786 - loss 0.20620940 - time (sec): 30.94 - samples/sec: 8015.46 - lr: 0.000011 - momentum: 0.000000
2023-10-19 21:02:40,204 ----------------------------------------------------------------------------------------------------
2023-10-19 21:02:40,205 EPOCH 8 done: loss 0.2060 - lr: 0.000011
2023-10-19 21:02:43,005 DEV : loss 0.19445763528347015 - f1-score (micro avg) 0.5412
2023-10-19 21:02:43,020 saving best model
2023-10-19 21:02:43,054 ----------------------------------------------------------------------------------------------------
2023-10-19 21:02:46,193 epoch 9 - iter 178/1786 - loss 0.20595102 - time (sec): 3.14 - samples/sec: 8098.61 - lr: 0.000011 - momentum: 0.000000
2023-10-19 21:02:49,523 epoch 9 - iter 356/1786 - loss 0.20915004 - time (sec): 6.47 - samples/sec: 7781.12 - lr: 0.000010 - momentum: 0.000000
2023-10-19 21:02:52,616 epoch 9 - iter 534/1786 - loss 0.21073170 - time (sec): 9.56 - samples/sec: 7757.88 - lr: 0.000009 - momentum: 0.000000
2023-10-19 21:02:55,656 epoch 9 - iter 712/1786 - loss 0.21153248 - time (sec): 12.60 - samples/sec: 7796.02 - lr: 0.000009 - momentum: 0.000000
2023-10-19 21:02:58,639 epoch 9 - iter 890/1786 - loss 0.20761614 - time (sec): 15.58 - samples/sec: 7833.49 - lr: 0.000008 - momentum: 0.000000
2023-10-19 21:03:01,755 epoch 9 - iter 1068/1786 - loss 0.20308539 - time (sec): 18.70 - samples/sec: 7955.21 - lr: 0.000008 - momentum: 0.000000
2023-10-19 21:03:04,756 epoch 9 - iter 1246/1786 - loss 0.20231833 - time (sec): 21.70 - samples/sec: 7957.84 - lr: 0.000007 - momentum: 0.000000
2023-10-19 21:03:08,025 epoch 9 - iter 1424/1786 - loss 0.20265135 - time (sec): 24.97 - samples/sec: 7928.87 - lr: 0.000007 - momentum: 0.000000
2023-10-19 21:03:11,119 epoch 9 - iter 1602/1786 - loss 0.20136577 - time (sec): 28.06 - samples/sec: 7934.50 - lr: 0.000006 - momentum: 0.000000
2023-10-19 21:03:14,180 epoch 9 - iter 1780/1786 - loss 0.19963572 - time (sec): 31.12 - samples/sec: 7968.46 - lr: 0.000006 - momentum: 0.000000
2023-10-19 21:03:14,278 ----------------------------------------------------------------------------------------------------
2023-10-19 21:03:14,279 EPOCH 9 done: loss 0.1998 - lr: 0.000006
2023-10-19 21:03:16,670 DEV : loss 0.19829225540161133 - f1-score (micro avg) 0.5422
2023-10-19 21:03:16,685 saving best model
2023-10-19 21:03:16,720 ----------------------------------------------------------------------------------------------------
2023-10-19 21:03:19,410 epoch 10 - iter 178/1786 - loss 0.18647044 - time (sec): 2.69 - samples/sec: 8782.65 - lr: 0.000005 - momentum: 0.000000
2023-10-19 21:03:22,378 epoch 10 - iter 356/1786 - loss 0.18609814 - time (sec): 5.66 - samples/sec: 8421.25 - lr: 0.000004 - momentum: 0.000000
2023-10-19 21:03:25,553 epoch 10 - iter 534/1786 - loss 0.18905060 - time (sec): 8.83 - samples/sec: 8418.27 - lr: 0.000004 - momentum: 0.000000
2023-10-19 21:03:28,473 epoch 10 - iter 712/1786 - loss 0.18325220 - time (sec): 11.75 - samples/sec: 8332.62 - lr: 0.000003 - momentum: 0.000000
2023-10-19 21:03:31,603 epoch 10 - iter 890/1786 - loss 0.18627484 - time (sec): 14.88 - samples/sec: 8203.30 - lr: 0.000003 - momentum: 0.000000
2023-10-19 21:03:34,723 epoch 10 - iter 1068/1786 - loss 0.18580625 - time (sec): 18.00 - samples/sec: 8209.34 - lr: 0.000002 - momentum: 0.000000
2023-10-19 21:03:37,964 epoch 10 - iter 1246/1786 - loss 0.18908200 - time (sec): 21.24 - samples/sec: 8087.53 - lr: 0.000002 - momentum: 0.000000
2023-10-19 21:03:41,184 epoch 10 - iter 1424/1786 - loss 0.19346145 - time (sec): 24.46 - samples/sec: 8079.92 - lr: 0.000001 - momentum: 0.000000
2023-10-19 21:03:44,560 epoch 10 - iter 1602/1786 - loss 0.19442863 - time (sec): 27.84 - samples/sec: 8041.93 - lr: 0.000001 - momentum: 0.000000
2023-10-19 21:03:47,563 epoch 10 - iter 1780/1786 - loss 0.19490021 - time (sec): 30.84 - samples/sec: 8043.92 - lr: 0.000000 - momentum: 0.000000
2023-10-19 21:03:47,661 ----------------------------------------------------------------------------------------------------
2023-10-19 21:03:47,661 EPOCH 10 done: loss 0.1948 - lr: 0.000000
2023-10-19 21:03:50,516 DEV : loss 0.19669239223003387 - f1-score (micro avg) 0.539
2023-10-19 21:03:50,560 ----------------------------------------------------------------------------------------------------
2023-10-19 21:03:50,560 Loading model from best epoch ...
2023-10-19 21:03:50,642 SequenceTagger predicts: Dictionary with 17 tags: O, S-PER, B-PER, E-PER, I-PER, S-LOC, B-LOC, E-LOC, I-LOC, S-ORG, B-ORG, E-ORG, I-ORG, S-HumanProd, B-HumanProd, E-HumanProd, I-HumanProd
2023-10-19 21:03:55,238
Results:
- F-score (micro) 0.4292
- F-score (macro) 0.2773
- Accuracy 0.2824
By class:
precision recall f1-score support
LOC 0.4196 0.5361 0.4707 1095
PER 0.4366 0.4931 0.4631 1012
ORG 0.1986 0.1569 0.1753 357
HumanProd 0.0000 0.0000 0.0000 33
micro avg 0.4044 0.4573 0.4292 2497
macro avg 0.2637 0.2965 0.2773 2497
weighted avg 0.3893 0.4573 0.4192 2497
2023-10-19 21:03:55,238 ----------------------------------------------------------------------------------------------------
|